← all playbooks

Get a one-page report on your desk from your own ATS every week

A short weekly report read straight from your ATS: live jobs with the candidates at each stage, what moved this week, the jobs with no recorded change, and a chase list. It says plainly what your system does not let it see.

use when
you want to start the week knowing which jobs are live, what moved, which jobs have gone quiet and who needs chasing, without clicking through every record
starts from
Your database

The prompt

Paste it into Claude Code or the Claude desktop app with Hyreflow connected. The first line tells your agent to use Hyreflow, so it reads the play, asks before it spends anything, and hands the work back to you.

paste this into Claude
Use hyreflow and give me a short weekly report on my desk from my own ATS.

My ATS is <ATS_NAME>. Cover the jobs owned by <OWNER_OR_TEAM>. Treat
these statuses as live: <LIVE_JOB_STATUSES>. Week starts <WEEK_START_DAY>.

- Live jobs by client, with how many candidates sit at each stage.
- What moved this week: candidates who changed stage, candidates added to
  a job, jobs opened and jobs closed.
- Jobs with no recorded change in <QUIET_DAYS> days, oldest first.
- A chase list: candidates who have sat at <STAGES_THAT_NEED_CHASING> for
  more than <STUCK_DAYS> days, and live jobs with nobody past the first
  stage.
- Tell me what you could not see in my ATS, so I do not read a gap as
  good news.

This is read-only. Do not write anything to the ATS and do not contact
anyone. Keep it to one page, with the numbers first.

Run it on five jobs first so I can check it against what I know, and
tell me what the full run costs before you do the rest. If it is right,
tell me what saving it as a weekly workflow would involve.

Replace every <PLACEHOLDER> with your own detail. Everything else can stay as written.

What you need first

  • A connected ATS: JobAdder, Atlas, Spott, Recruit CRM, Recruiterflow or Loxo
  • Jobs and pipeline stages kept up to date in it, because the report can only read what was recorded
  • Your own rules: which job statuses count as live, and how many days without a change make a job quiet
  • A Hyreflow workspace. Reading your own ATS uses no credits, the written summary of a scheduled run does

Tools it can reach for

The agent picks per step from what your workspace has. Nothing here is required by name.

What happens when you run it

Free steps are marked free. Anything that spends credits is marked, and the agent asks before the first paid run of any size.

  1. 1

    Agree what the report covers

    free

    Whose jobs, which statuses count as live, where the week starts, and how many days without a change make a job quiet. A scheduled run cannot ask you questions, so the rules are settled here.

  2. 2

    Read the live jobs

    free

    Jobs are read from your ATS with the status, client and owner the record carries, and narrowed to the live ones. Where a system does not return the status you work by, you name the live jobs. It is your own data on your own account, so the read uses no credits.

  3. 3

    Read the pipeline on each job

    free

    For each live job, the candidates attached to it and the stage each one sits at, counted by stage. Recruit CRM is the exception: it gives stages one candidate at a time, so the stage picture there is built from the candidates whose records changed this week.

  4. 4

    Find what moved this week

    free

    JobAdder, Atlas, Spott and Recruit CRM can be asked for the jobs and the candidate records changed since a date, which is what makes a weekly view possible. Recruiterflow and Loxo give the pipeline as it stands today, so on those the report is a snapshot and says so.

  5. 5

    Flag the quiet jobs and build the chase list

    free

    Live jobs with no dated change inside your limit are listed oldest first. Candidates who have sat too long at the stages you named go on the chase list with the job, the client and the date of the last change.

  6. 6

    Write the one-page report

    credits

    The numbers are turned into a short report with the totals first and the exceptions under them. In conversation your assistant writes it. In a scheduled run there is no assistant, so this is a metered AI step and the only part of the play that uses credits.

  7. 7

    Save it as a weekly workflow

    free

    Once a hand run matches what you know about your desk, the same reads are saved as a workflow on a weekly schedule. Each run leaves the report in your workspace. It sends nothing and writes nothing to the ATS.

What a run costs

Nothing in this play spends credits.

Free before anything is charged

  • Agree what the report covers
  • Read the live jobs
  • Read the pipeline on each job
  • Find what moved this week
  • Flag the quiet jobs and build the chase list
  • Save it as a weekly workflow

What moves the number

  • How many roles survive the free filters. Everything dropped before the paid steps costs nothing.
  • The scoring and drafting steps run on the metered agent, charged on what they read and write rather than per role, so they sit outside this table.
  • Providers you connect with your own key. Those calls bill your account, not your credits.

An estimate, not a quote, priced at the volume credit rate. Your agent sizes the run against your own workspace and tells you what it will cost before it spends anything.

The report you write from memory is already in your ATS

Most desks review the week the same way. Somebody asks how the jobs are going, and you answer from memory: the three roles you worked hardest, the offer that is close, the client who went silent. The jobs you did not think about do not come up. Those are the ones that needed the question.

Your ATS holds the whole picture. It knows every live job, who is at which stage, and which records changed. What it does not do is read across all of it and tell you, on one page, where the week went and what has stalled. That is a reading job, and reading your own system costs nothing.

The report is short on purpose. Totals first, then the exceptions: what moved, what did not, who is waiting on you. A page you read beats a dashboard you scroll past.

What you get back

One page. At the top, the counts: live jobs, candidates in process by stage, jobs opened and closed this week. Then three lists. What moved, with the candidate, the job and the stage they moved to. The quiet jobs, oldest first, with the date of the last recorded change. The chase list, with the candidate, the client and the date their record last changed.

At the foot, a plain note of what the report could not see in your system, so an empty section is never mistaken for a clean one.

Variations worth knowing

One report per consultant. Filter by job owner and each person gets their own page.

Client by client. Group by client before an account review. It shows whose roles are ageing without movement.

Turn a quiet job into action. A live role with a thin pipeline is a prompt to look again at who you already hold: find the candidates already sitting in your own ATS.

Fix what it exposes. If the report is wrong because the records are, that is the finding. Make an old recruitment database searchable again is the follow-up.

Where this goes wrong

Stages nobody updates. The report is as true as the pipeline. If candidates are moved on in conversation and not in the system, the chase list fills with people who were dealt with last week.

Reading quiet as idle. A job worked entirely by phone shows no recorded change, which is why the report uses those words.

A snapshot read as a trend. On systems that cannot answer what changed since a date, the report shows today. It does not show movement, and it does not pretend to.

Live statuses set wrong. Leave out a status your team uses and those jobs vanish from the report. Check the first run against jobs you know.

A report nobody opens. A scheduled report waits in your workspace. If opening it is not part of your Monday, it changes nothing.

Questions

Which systems does this work with?

JobAdder, Atlas, Spott, Recruit CRM, Recruiterflow and Loxo, because those connections can read jobs together with pipeline or stage data. The first four also answer what changed since a date. Recruiterflow and Loxo return the pipeline as it stands, so the report on those is a snapshot of today. For any other system, export your jobs and pipeline to a file and the same report is built from that.

What can it not see?

Anything that was never recorded as a change on a job, an application or a candidate. Phone calls, emails and interviews held in your calendar are invisible unless they moved a stage or touched the record. No connection returns a last-contacted date. So a quiet job means no recorded change, not no work, and the report uses those words.

Does it cost anything?

Reading your own ATS uses no Hyreflow credits, and most of this play is reading, counting and filtering. Run by hand, the summary is written by the assistant you are already talking to. Saved as a scheduled workflow, the written summary is a metered AI step, and the hand run shows you what one run costs before any schedule exists.

Can it run every Monday without me?

Yes, once you have run it by hand and checked it against jobs you know well. It converts into a scheduled workflow on a weekly trigger, and each run leaves the report in your workspace for you to open. Prove it by hand first, because a workflow applies your rules exactly as written: a wrong list of live statuses produces a confident, wrong report every week.

Will it change anything in my ATS?

No. Every call in this play is a read. It does not move candidates, close jobs, write notes or message anyone. If the report shows records that need fixing, that is a separate job with its own review.